Beanstalk Longshot test farm
Menu

Report rp_f6a4e2853ae4d979 fail

#2004 WIP: perf(vk): a compute program's c0 layout kept by each thread between the words it reads (RX_VK_CS_LAYOUT_FAST)

any on manasuazo/agent-03253d,

This report disagrees with its peers.

kokiri-walk forest.frametime_1pct_low_ms: -10.1% against peers' +0.7%
kokiri-walk forest.frametime_p99_ms: -16.6% against peers' +0.8%
kokiri-walk forest.hitch_count: -38.9% against peers' +3500.0%
kokiri-walk forest.stutter_count: -50.0% against peers' +6000.0%
kokiri-walk#1 forest.frametime_1pct_low_ms: -20.2% against peers' +2.1%
kokiri-walk#1 forest.frametime_max_ms: +8.9% against peers' -3.5%
kokiri-walk#1 forest.frametime_p99_ms: -25.3% against peers' +0.9%
kokiri-walk#1 run_hitch_count: -22.2% against peers' -3.7%

Machine

Commit
614cc3ee060a1523a896d74b52bb5428d3cb7724
Baseline
9dd41326e71094d8b774459dac5154eb5a3306d9
App
nightly-221-a9505cc3, attempt 1
Game build
e0f3ccfa
os
linux
cpu
AMD Ryzen 5 5500F 6-Core Processor
gpu
AMD Radeon RX 580 Series (RADV POLARIS10)
tags
linuxamdradvdgpu
driver
26.2.7
ram_gb
32
vram_mb
8,192
gpu_vendor
amd
os_version
<host> (kernel 7.3.4)

Builds

main (baseline)

0 s warm

from 9dd41326e710

the base was built already

PR

43 s warm

Program cache: warm

Scenarios

kokiri-walk RX_VK_CS_LAYOUT_FAST=0 (PR only)

ok
Metric (median)mainPRΔ
exit_code00±0%
run_hitch_count7364−12.3%
forest.hitch_count3622−38.9%
log_forbidden_hits00±0%
forest.stutter_count2613−50%
log_expected_missing00±0%
run_frametime_max_ms405.05446.82+10.3%
forest.frametime_max_ms95.5585.04−11%
forest.frametime_p50_ms27.9428.24+1.1%
forest.frametime_p99_ms60.1650.15−16.6%
forest.frametime_1pct_low_ms70.563.38−10.1%
screenshot_diff—0.84—

Pictures against main: forest diff 0.8358 (main against itself )

Runs (2)
Metricmain 0pr 1
hangnono
crashednono
exit_code00
run_hitch_count7364
forest.hitch_count3622
log_forbidden_hits00
forest.stutter_count2613
log_expected_missing00
run_frametime_max_ms405.05446.82
forest.frametime_max_ms95.5585.04
forest.frametime_p50_ms27.9428.24
forest.frametime_p99_ms60.1650.15
forest.frametime_1pct_low_ms70.563.38
screenshot_diff—0.84

kokiri-walk RX_VK_CS_LAYOUT_FAST=1 (PR only)

ok
Metric (median)mainPRΔ
exit_code00±0%
run_hitch_count6349−22.2%
forest.hitch_count287−75%
log_forbidden_hits00±0%
forest.stutter_count185−72.2%
log_expected_missing00±0%
run_frametime_max_ms466.59478.67+2.6%
forest.frametime_max_ms78.3185.28+8.9%
forest.frametime_p50_ms28.1928.73+1.9%
forest.frametime_p99_ms53.7740.17−25.3%
forest.frametime_1pct_low_ms66.2652.9−20.2%
screenshot_diff—0.85—

Pictures against main: forest diff 0.8533 (main against itself )

Runs (2)
Metricmain 0pr 1
hangnono
crashednono
exit_code00
run_hitch_count6349
forest.hitch_count287
log_forbidden_hits00
forest.stutter_count185
log_expected_missing00
run_frametime_max_ms466.59478.67
forest.frametime_max_ms78.3185.28
forest.frametime_p50_ms28.1928.73
forest.frametime_p99_ms53.7740.17
forest.frametime_1pct_low_ms66.2652.9
screenshot_diff—0.85

Rules

ScenarioCellRuleLimitmainPRΔ %Status
kokiri-walkRX_VK_CS_LAYOUT_FAST=0 (PR only)crashFalse00—pass
kokiri-walkRX_VK_CS_LAYOUT_FAST=0 (PR only)screenshot.max_diff0.02—0.84—fail
kokiri-walkRX_VK_CS_LAYOUT_FAST=0 (PR only)forest.frametime_p99_ms.max_delta_pct1060.1650.15−16.6%inconclusiveneed two usable starts per side (baseline: 1, PR: 1)
kokiri-walkRX_VK_CS_LAYOUT_FAST=0 (PR only)log_forbidden_hits.max000—pass
kokiri-walkRX_VK_CS_LAYOUT_FAST=1 (PR only)crashFalse00—pass
kokiri-walkRX_VK_CS_LAYOUT_FAST=1 (PR only)screenshot.max_diff0.02—0.85—fail
kokiri-walkRX_VK_CS_LAYOUT_FAST=1 (PR only)forest.frametime_p99_ms.max_delta_pct1053.7740.17−25.3%inconclusiveneed two usable starts per side (baseline: 1, PR: 1)
kokiri-walkRX_VK_CS_LAYOUT_FAST=1 (PR only)log_forbidden_hits.max000—pass
kokiri-walkRX_VK_CS_LAYOUT_FAST=1 (PR only)forest.frametime_p99_ms: RX_VK_CS_LAYOUT_FAST=1 vs RX_VK_CS_LAYOUT_FAST=0 (PR)550.1540.17−19.9%passthe PR's own cells: RX_VK_CS_LAYOUT_FAST=1 against RX_VK_CS_LAYOUT_FAST=0 (RX_VK_CS_LAYOUT_FAST=0 (PR only))

Pictures of the failure

The last frame of a step that got stuck, or main's and the PR's picture where they differed more than the plan allows; shrunk, and kept a day: until .

kokiri-walk-c0-forest-main-failed.png
kokiri-walk-c0-forest-main-failed.png
kokiri-walk-c0-forest-pr-failed.png
kokiri-walk-c0-forest-pr-failed.png
kokiri-walk-c1-forest-main-failed.png
kokiri-walk-c1-forest-main-failed.png

Log (PR side, end)

[hos host] vk: host time in GPU resource waits: 16.24 ms a frame (fences 0.04, timelines 16.20, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 643.2 full barriers recorded, 313.8 left out (nothing recorded since the one before), 17.2 pass starts ordered in their fragment stages only (draw passes after draw passes)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 53.6; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: a frame: indirect mesh draws' arguments (RX_VK_MESH_CONV=2): 41.0 converted one by one, 0.0 from 0.0 batches at the game's barriers (0.0 converted in them; 0.0 dropped after a write, 0.0 the ring came round), 41.0 for the compute fallback (no mesh shaders in use: one by one); scans ended at a barrier 0.0, an op 0.0, the end 0.0, full 0.0; the draws batched by a scan right after an op that ended the one before, by op: none
[hos host] vk: GPU time 21.8 ms a frame (the longest submission 6.3 ms); 0.0 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.0 MB), ASTC or BC 0.0 (0.0 MB), other 0.0 (0.0 MB)
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos host] vk: queues ordered by their events: 21618 submissions stopped at a wait, 21572 went on once it was reached, 46 waits gone past; made to go by presents 0, releases 0, waits for the replay 0, 50 ms 4 (stopped 50.06 ms at most); waits met 55032, not met 0; captures at a CPU event's wait 3051, 0 of them at a wait met on arrival
[hos host] vk: the render thread (600 presents, 28.59 ms a frame): its jobs 11.76 ms a frame (submits 1.12, presents 0.16, others 10.48), idle 16.83; the game waited for room at 0 presents, 0.00 ms each (0.00 a frame; 1.00 jobs queued at a present); 0 catch-ups (a finish, a texture read back, a write-back not recorded yet: 0), 0.00 ms each; 0 presents held for the CPU's signal, 0.00 ms each (0 let go by RX_VK_HOLD_MS); 599 captures at a CPU signal; the submits' waits read in 0.036 ms a frame; readbacks: 3000 next to a texture's storage (no catch-up), 0 in one, 0 read exactly (after the work handed over: among the catch-ups); a present made 10.52 ms after its hand-over (11.97 at most)
[hos host] vk: frame overlap (600 presents): the GPU between recordings, ms a frame, idle + in the next one's uploads: up to a frame's first recording with a draw 0.11 + 0.90 (the window's copy is in the idle), after it 0.87 + 4.99; after a frame's last recording that reads or writes the game's memory come 1.00 recordings (0.22 ms of GPU time), after its last that reads it 1.00 (0.22 ms)
[hos host] vk: frame ahead (600 presents): 9.00 parts a frame recorded in the signal's job, 66.2% of them took nothing that came with the signal; of the others, 3.04 a frame took pages of the signal's capture into a mirror (132.8 pages a frame) and 0.00 pages read as guest memory had them (0.0 pages a frame); a frame's first part took nothing in 0 of 599 frames, its first three in 0
[hos host] vk: frame ahead, what came with the signal: 599 signals with a capture, each of 164.8 pages in 10.0 pools; between a frame's last submit and that signal 0.00 texture and sampler registrations, 0.00 releases and 0.00 other CPU signals a signal (0 of them with any)
[hos host] vk: frame overlap, by place: idle + uploads before a frame's 2nd to 12th recording with a draw, then all later ones: 0.75+2.33, 0.02+0.35, 0.02+1.79, 0.01+0.08, 0.02+0.09, 0.01+0.08, 0.02+0.11, 0.01+0.08, 0.01+0.08, 0.00+0.00, 0.00+0.00, 0.00+0.00
[hos host] vk: replay reads of guest memory as submitted: 2140000 (867384 pages from the captures)
[hos host] vk: render targets drawn over from another start, let go: 318 so far (474 MB), 307 of them (457 MB) between the passes
[hos BaseProcExecutor1] vcores: 62986 takes, 7997 waits (mean 73 us, longest 4.5 ms, 0 over HOS_VCORE_LOG_MS), 3207 kicks, 3503 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: texture slots registered again at another size: 200000 (of 278540 registrations; slot 30659 256x256 -> 30x16)
[hos host] vk: replay reads of guest memory as submitted: 2160000 (874933 pages from the captures)
[hos host] vk: write-backs left out (600 submits): 814, 3411.9 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 4936 dirty pages from their submits' captures, 46706 from earlier ones', 0 from guest memory as it was at the replay; 24 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 865 dispatches, 37500 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5944 runs, 636.5 MB compared; 75 pages read into the shadows; 256 MB of shadows (121 MB of pages known, in 154 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 18137 in captures, 0 only in the collection, 38 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 2180000 (884543 pages from the captures)
[hos ModuleSystemWorker4] vk: readbacks of GPU results: 980 of 3000 waited for their write-back, 0.75 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos ModuleSystemWorker4] vk: invalidates since the last readbacks line: 3742 kept their virtual core (nothing there to wait for), 3170 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos Presentation Thread] vi: 3639 frames presented (34.9 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 0, 15.5-17.8 0, 17.8-22 0, 22-30 169, 30-40 6, over 40 0
[hos host] vk: replay reads of guest memory as submitted: 2200000 (892209 pages from the captures)
[hos ModuleSystemWorker2] sync: object cache (HOS_OBJ_CACHE=16384 a thread, 33 made): 9145066 lookups, 76928 missed (0.84 %)
[hos host] vk: replay reads of guest memory as submitted: 2220000 (899705 pages from the captures)
[hos host] vk: sparse mirrors: 49, 797 MB with memory of 964 MB (166 MB spared); blocks given memory later 60, given back 1234; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 90 binds (14.2 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 7 taken from the spares, 60 made new; 18 kept as spares, 0 destroyed; 11 spares now, 296 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 836, 3504.2 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 5080 dirty pages from their submits' captures, 47616 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 885 dispatches, 38296 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6105 runs, 654.0 MB compared; 75 pages read into the shadows; 256 MB of shadows (121 MB of pages known, in 153 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 18598 in captures, 0 only in the collection, 37 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 58 live (1088 MB), 0 retired still imported (0 MB), 8 retired given back (236 MB)
[hos host] vk: program picks: 2497464 kept (1328176 by the last picks' textures, a uniform buffer bound since), 19746 new, 1082789 with other textures (of 3600000)
[hos Presentation Thread] vk: capture: a submit 0.12 ms (lock 0.02, scan 0.03 over 1309 ranges, 1868 MB), 215 pages written, 215 copied (0.8 MB)
[hos host] vk: a capture taken (ms): all 0.016; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.002, bytes copied 0.000; 24 kept (128 MB), 12 spare buffers, 0 texture pages copied (0 MB)
[hos sead::MainThread] nvn: CommandBufferSignalEvent #40000: event 0x100d6c8038 mode 0 value 39999 (lr 0x8066189c)
[hos host] vk: replay reads of guest memory as submitted: 2240000 (907337 pages from the captures)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.02, writes collected 0.03
[hos host] vk: replay reads of guest memory as submitted: 2260000 (915044 pages from the captures)
[hos host] vk: the recording streams: 48 of 1 MB, the most one held 176 KB; 0 made on demand (0 MB kept now); c1 stages: at most 121 a recording; 0 descriptor pools made when a recording's own ran out (0 slots' pools made whole); a recording's pool took at most 138 sets (ub 1647, sb 917, cis 9267, si 0, utb 0, stb 0)
[hos host] vk: write-backs left out (600 submits): 814, 3411.9 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 4881 dirty pages from their submits' captures, 46251 from earlier ones', 0 from guest memory as it was at the replay; 24 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 863 dispatches, 37167 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5945 runs, 635.1 MB compared; 75 pages read into the shadows; 256 MB of shadows (121 MB of pages known, in 154 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 18139 in captures, 0 only in the collection, 38 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 2280000 (922717 pages from the captures)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 0 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: a frame: 60 draws kept their neighbour's descriptors (0 of them through other slots' registrations)
[hos host] vk: a frame: 0.0 draws left out (not ready), 0.00 waits for shaders or pipelines (0.0 ms); translation 3676 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 10.86, present 0.16, draw 6.66, descriptors 2.42, texture lookups 0.08, uploads 0.00, pipelines 0.19, programs 1.09, commands 2.56, dispatch 0.86, memory sync 0.32, queue submit 1.28, slot wait 0.07
[hos host] vk: a frame: mirrors 67 (797 MB): 5.42 MB synced in 1005.3 copies, 17.19 MB written back
[hos host] vk: 3720 presents, 34.9 fps; a frame: 1775 draws (0 skipped), 139 dispatches, 170 passes, 152 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.1 ms (recording slots only); 1090 pipelines (486 ms), 660 programs (11 failed), 991 textures (3554 MB)
[hos host] vk: host time in GPU resource waits: 16.25 ms a frame (fences 0.04, timelines 16.22, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 643.5 full barriers recorded, 313.8 left out (nothing recorded since the one before), 17.2 pass starts ordered in their fragment stages only (draw passes after draw passes)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 53.6; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: a frame: indirect mesh draws' arguments (RX_VK_MESH_CONV=2): 41.0 converted one by one, 0.0 from 0.0 batches at the game's barriers (0.0 converted in them; 0.0 dropped after a write, 0.0 the ring came round), 41.0 for the compute fallback (no mesh shaders in use: one by one); scans ended at a barrier 0.0, an op 0.0, the end 0.0, full 0.0; the draws batched by a scan right after an op that ended the one before, by op: none
[hos host] vk: GPU time 21.8 ms a frame (the longest submission 6.3 ms); 0.0 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.0 MB), ASTC or BC 0.0 (0.0 MB), other 0.0 (0.0 MB)
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos host] vk: late pages found at the replay (600 takes): 0 pages in 0 takes (RX_VK_LATE_AT_REPLAY=0: only the ones the capture found)
[hos host] vk: replay reads of guest memory as submitted: 2300000 (930328 pages from the captures)
[hos host] vk: sparse mirrors: 49, 797 MB with memory of 964 MB (166 MB spared); blocks given memory later 60, given back 1234; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 90 binds (14.2 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 7 taken from the spares, 60 made new; 18 kept as spares, 0 destroyed; 11 spares now, 296 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 836, 3504.2 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 5057 dirty pages from their submits' captures, 47661 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 890 dispatches, 38304 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6085 runs, 651.2 MB compared; 75 pages read into the shadows; 256 MB of shadows (121 MB of pages known, in 153 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 18622 in captures, 0 only in the collection, 37 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 58 live (1088 MB), 0 retired still imported (0 MB), 8 retired given back (236 MB)
[hos sead::MainThread] nvn: GPU waits in 5 s: 175 sync waits (1926 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos host] vk: replay reads of guest memory as submitted: 2320000 (938776 pages from the captures)
[hos host] vk: readback pages refreshed from guest memory: 802477 pages so far (0 held: may be written ahead in their submission)
[hos Presentation Thread] vk: capture: a submit 0.10 ms (lock 0.00, scan 0.02 over 1308 ranges, 1864 MB), 216 pages written, 216 copied (0.8 MB)
[hos host] vk: a capture taken (ms): all 0.016; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.002, bytes copied 0.000; 24 kept (124 MB), 12 spare buffers, 0 texture pages copied (0 MB)
[hos ModuleSystemWorker2] barrier: 53763 looks: 20995 found the masks moved already, 31677 waited until they moved, 1091 for the whole 1 ms (a wait 316 us on average); 25441 wakes
[hos host] vk: replay reads of guest memory as submitted: 2340000 (946587 pages from the captures)
[hos ModuleSystemWorker1] vk: readbacks of GPU results: 980 of 3000 waited for their write-back, 0.77 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos ModuleSystemWorker1] vk: invalidates since the last readbacks line: 3654 kept their virtual core (nothing there to wait for), 3002 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: write-backs left out (600 submits): 836, 3504.2 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 4995 dirty pages from their submits' captures, 47564 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 882 dispatches, 38391 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6071 runs, 651.7 MB compared; 77 pages read into the shadows; 256 MB of shadows (121 MB of pages known, in 154 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 18632 in captures, 0 only in the collection, 39 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 2360000 (955011 pages from the captures)
[hos host] vk: program picks: 2637499 kept (1402602 by the last picks' textures, a uniform buffer bound since), 19856 new, 1142644 with other textures (of 3800000)
[hos AudioFrameworkThread] vcores: 62016 takes, 8216 waits (mean 81 us, longest 4.8 ms, 0 over HOS_VCORE_LOG_MS), 3260 kicks, 3635 yields, 0 starved grants, 0 escapes in 5 s
[hos sead::MainThread] nvn: CommandBufferWaitEvent #60000: event 0x1006e38000 mode 0 value 1 (now 0) (lr 0x802eb444)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.02, writes collected 0.03
[hos host] vk: replay reads of guest memory as submitted: 2380000 (962780 pages from the captures)
[hos rmbl::VibrationThread] hid: 16384 vibration values, strongest 0.00 since the last count
[hos Presentation Thread] vi: 3813 frames presented (34.7 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 0, 15.5-17.8 0, 17.8-22 0, 22-30 164, 30-40 10, over 40 0
[hos host] vk: slow submit: 51.9 ms: draw 4.7 descriptors 1.8 commands 1.6
[hos host] vk: replay reads of guest memory as submitted: 2400000 (971538 pages from the captures)
[hos host] vk: sparse mirrors: 49, 797 MB with memory of 964 MB (166 MB spared); blocks given memory later 60, given back 1234; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 90 binds (14.2 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 7 taken from the spares, 60 made new; 18 kept as spares, 0 destroyed; 11 spares now, 296 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 836, 3504.2 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 5057 dirty pages from their submits' captures, 47568 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 886 dispatches, 38376 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6051 runs, 650.6 MB compared; 76 pages read into the shadows; 256 MB of shadows (121 MB of pages known, in 153 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 18662 in captures, 0 only in the collection, 37 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 58 live (1088 MB), 0 retired still imported (0 MB), 8 retired given back (236 MB)
[hos host] vk: replay reads of guest memory as submitted: 2420000 (979476 pages from the captures)
[hos host] vk: a frame: 60 draws kept their neighbour's descriptors (0 of them through other slots' registrations)
[hos host] vk: a frame: 0.0 draws left out (not ready), 0.00 waits for shaders or pipelines (0.0 ms); translation 3676 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 11.22, present 0.16, draw 6.48, descriptors 2.31, texture lookups 0.09, uploads 0.00, pipelines 0.19, programs 1.05, commands 2.54, dispatch 0.87, memory sync 0.30, queue submit 1.31, slot wait 0.07
[hos host] vk: a frame: mirrors 67 (797 MB): 5.41 MB synced in 1007.5 copies, 17.14 MB written back
[hos host] vk: 3840 presents, 33.8 fps; a frame: 1734 draws (0 skipped), 140 dispatches, 170 passes, 153 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.1 ms (recording slots only); 1090 pipelines (486 ms), 660 programs (11 failed), 991 textures (3554 MB)
[hos host] vk: host time in GPU resource waits: 17.22 ms a frame (fences 0.04, timelines 17.18, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 643.0 full barriers recorded, 314.1 left out (nothing recorded since the one before), 17.2 pass starts ordered in their fragment stages only (draw passes after draw passes)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 54.3; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: a frame: indirect mesh draws' arguments (RX_VK_MESH_CONV=2): 41.0 converted one by one, 0.0 from 0.0 batches at the game's barriers (0.0 converted in them; 0.0 dropped after a write, 0.0 the ring came round), 41.0 for the compute fallback (no mesh shaders in use: one by one); scans ended at a barrier 0.0, an op 0.0, the end 0.0, full 0.0; the draws batched by a scan right after an op that ended the one before, by op: none
[hos host] vk: GPU time 22.0 ms a frame (the longest submission 6.3 ms); 0.0 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.0 MB), ASTC or BC 0.0 (0.0 MB), other 0.0 (0.0 MB)
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos sead::MainThread] vk: capture: a submit 0.10 ms (lock 0.00, scan 0.02 over 1308 ranges, 1864 MB), 216 pages written, 216 copied (0.8 MB)
[hos host] vk: a capture taken (ms): all 0.017; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.002, bytes copied 0.000; 25 kept (138 MB), 10 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: replay reads of guest memory as submitted: 2440000 (987696 pages from the captures)
[hos host] vk: write-backs left out (600 submits): 836, 3504.2 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 5022 dirty pages from their submits' captures, 47598 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 883 dispatches, 38357 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5999 runs, 649.1 MB compared; 76 pages read into the shadows; 256 MB of shadows (121 MB of pages known, in 153 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 18664 in captures, 0 only in the collection, 38 only by the replay itself
[hos host] vk:   read live, by pool:
[hos ModuleSystemWorker3] nvn: 400000 recordings made, 49035 live copies (16.7 MB), 370 texture copies; 2 command buffers' segments (4.1 MB)
[hos host] vk: replay reads of guest memory as submitted: 2460000 (996401 pages from the captures)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 0 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: late pages found at the replay (600 takes): 0 pages in 0 takes (RX_VK_LATE_AT_REPLAY=0: only the ones the capture found)
[hos host] vk: slow submit: 35.7 ms: draw 1.2
[hos host] vk: replay reads of guest memory as submitted: 2480000 (1003946 pages from the captures)
[hos host] vk: the recording streams: 48 of 1 MB, the most one held 176 KB; 0 made on demand (0 MB kept now); c1 stages: at most 121 a recording; 0 descriptor pools made when a recording's own ran out (0 slots' pools made whole); a recording's pool took at most 138 sets (ub 1647, sb 917, cis 9267, si 0, utb 0, stb 0)
[hos host] vk: sparse mirrors: 49, 797 MB with memory of 964 MB (166 MB spared); blocks given memory later 60, given back 1234; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 90 binds (14.2 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 7 taken from the spares, 60 made new; 18 kept as spares, 0 destroyed; 11 spares now, 296 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 814, 3411.9 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 5010 dirty pages from their submits' captures, 46509 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 869 dispatches, 37437 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6039 runs, 637.9 MB compared; 102 pages read into the shadows; 256 MB of shadows (121 MB of pages known, in 154 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 17891 in captures, 0 only in the collection, 38 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 58 live (1088 MB), 0 retired still imported (0 MB), 8 retired given back (236 MB)
[hos host] vk: program picks: 2776930 kept (1476354 by the last picks' textures, a uniform buffer bound since), 19966 new, 1203103 with other textures (of 4000000)
[hos host] vk: queues ordered by their events: 24000 submissions stopped at a wait, 23952 went on once it was reached, 46 waits gone past; made to go by presents 0, releases 0, waits for the replay 0, 50 ms 4 (stopped 50.06 ms at most); waits met 62466, not met 0; captures at a CPU event's wait 3349, 0 of them at a wait met on arrival
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.02, writes collected 0.03
[hos host] vk: replay reads of guest memory as submitted: 2500000 (1012649 pages from the captures)
[hos ModuleSystemWorker2] vk: readbacks of GPU results: 973 of 3000 waited for their write-back, 0.79 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos ModuleSystemWorker2] vk: invalidates since the last readbacks line: 3527 kept their virtual core (nothing there to wait for), 2873 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos sead::MainThread] nvn: GPU waits in 5 s: 169 sync waits (1986 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos host] vk: replay reads of guest memory as submitted: 2520000 (1020422 pages from the captures)
[hos host] vk: write-backs left out (600 submits): 836, 3504.2 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 5044 dirty pages from their submits' captures, 47800 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 891 dispatches, 38403 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6032 runs, 651.5 MB compared; 75 pages read into the shadows; 256 MB of shadows (121 MB of pages known, in 153 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 18620 in captures, 0 only in the collection, 37 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 2540000 (1029918 pages from the captures)
[hos sead::MainThread] vk: capture: a submit 0.10 ms (lock 0.00, scan 0.02 over 1308 ranges, 1864 MB), 215 pages written, 215 copied (0.8 MB)
[hos host] vk: a capture taken (ms): all 0.016; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.002, bytes copied 0.000; 25 kept (132 MB), 11 spare buffers, 0 texture pages copied (0 MB)
[hos sead::MainThread] nvn: CommandBufferSignalEvent #45000: event 0x100d6c8044 mode 0 value 44999 (lr 0x802742f0)
[hos host] vk: readback pages refreshed from guest memory: 875581 pages so far (0 held: may be written ahead in their submission)
[hos sead::MainThread] nvn: DrawMeshTasksIndirect: 100000 so far (x1 0x11789a4000 x2 0x1 x3 0x11789a4000 x4 0x4040)
[hos host] vk: replay reads of guest memory as submitted: 2560000 (1037778 pages from the captures)
[hos ModuleSystemWorker3] vcores: 61403 takes, 8166 waits (mean 70 us, longest 2.6 ms, 0 over HOS_VCORE_LOG_MS), 3355 kicks, 3676 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: a frame: 60 draws kept their neighbour's descriptors (0 of them through other slots' registrations)
[hos host] vk: a frame: 0.0 draws left out (not ready), 0.00 waits for shaders or pipelines (0.0 ms); translation 3676 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 11.14, present 0.18, draw 6.56, descriptors 2.36, texture lookups 0.08, uploads 0.00, pipelines 0.19, programs 1.08, commands 2.53, dispatch 0.88, memory sync 0.30, queue submit 1.30, slot wait 0.07
[hos host] vk: a frame: mirrors 67 (797 MB): 5.43 MB synced in 1007.4 copies, 17.18 MB written back
[hos host] vk: 3960 presents, 34.2 fps; a frame: 1766 draws (0 skipped), 140 dispatches, 170 passes, 153 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.1 ms (recording slots only); 1090 pipelines (486 ms), 660 programs (11 failed), 991 textures (3554 MB)
[hos host] vk: host time in GPU resource waits: 16.74 ms a frame (fences 0.04, timelines 16.70, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 643.3 full barriers recorded, 314.1 left out (nothing recorded since the one before), 17.2 pass starts ordered in their fragment stages only (draw passes after draw passes)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 54.1; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: a frame: indirect mesh draws' arguments (RX_VK_MESH_CONV=2): 41.0 converted one by one, 0.0 from 0.0 batches at the game's barriers (0.0 converted in them; 0.0 dropped after a write, 0.0 the ring came round), 41.0 for the compute fallback (no mesh shaders in use: one by one); scans ended at a barrier 0.0, an op 0.0, the end 0.0, full 0.0; the draws batched by a scan right after an op that ended the one before, by op: none
[hos host] vk: GPU time 21.9 ms a frame (the longest submission 6.5 ms); 0.0 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.0 MB), ASTC or BC 0.0 (0.0 MB), other 0.0 (0.0 MB)
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos host] vk: replay reads of guest memory as submitted: 2580000 (1045343 pages from the captures)
[hos host] vk: sparse mirrors: 49, 797 MB with memory of 964 MB (166 MB spared); blocks given memory later 60, given back 1234; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 90 binds (14.2 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 7 taken from the spares, 60 made new; 18 kept as spares, 0 destroyed; 11 spares now, 296 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 814, 3411.9 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 4932 dirty pages from their submits' captures, 46372 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 867 dispatches, 37307 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6044 runs, 638.9 MB compared; 75 pages read into the shadows; 256 MB of shadows (121 MB of pages known, in 154 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 18192 in captures, 0 only in the collection, 38 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 58 live (1088 MB), 0 retired still imported (0 MB), 8 retired given back (236 MB)
[hos Presentation Thread] vi: 3983 frames presented (33.8 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 0, 15.5-17.8 0, 17.8-22 0, 22-30 149, 30-40 19, over 40 2
[hos host] vk: replay reads of guest memory as submitted: 2600000 (1053039 pages from the captures)
[hos sead::MainThread] nvn: CommandBufferWaitEvent #65000: event 0x1006e38000 mode 0 value 1 (now 0) (lr 0x802eb444)
[hos ModuleSystemWorker5] sync: object cache (HOS_OBJ_CACHE=16384 a thread, 33 made): 8864516 lookups, 75742 missed (0.85 %)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.02, writes collected 0.03
[hos host] vk: replay reads of guest memory as submitted: 2620000 (1060755 pages from the captures)
[hos host] vk: program picks: 2916999 kept (1550724 by the last picks' textures, a uniform buffer bound since), 20075 new, 1262925 with other textures (of 4200000)
[hos host] vk: write-backs left out (600 submits): 836, 3504.2 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 5054 dirty pages from their submits' captures, 47615 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 884 dispatches, 38325 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6021 runs, 649.9 MB compared; 75 pages read into the shadows; 256 MB of shadows (121 MB of pages known, in 153 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 18694 in captures, 0 only in the collection, 37 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 2640000 (1068391 pages from the captures)
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 0 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos Presentation Thread] vk: capture: a submit 0.10 ms (lock 0.00, scan 0.02 over 1308 ranges, 1864 MB), 216 pages written, 216 copied (0.8 MB)
[hos host] vk: late pages found at the replay (600 takes): 120 pages in 1 takes (RX_VK_LATE_AT_REPLAY=0: only the ones the capture found)
[hos host] vk: a capture taken (ms): all 0.016; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.002, bytes copied 0.000; 24 kept (130 MB), 12 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: late pages' fix (600 recordings with one): of 600 captures with late pages 600 fixed (117.8 pages, 118.8 fixes a capture), waited as before: 0 no pool, 0 no stamps or shadow yet, 0 stamps too new, 0 whole copies, 0 in place or unknown, 0 a page's earlier fix not recorded yet, 0 other; 545 read on the CPU first, 0 with no table, 0 pages whose shadow changed first; stamps 390 blocks (24960 KB); of the reads on the CPU, 0 waited for an earlier part's fix, 0 copied the page again instead (that part not waited for); the paths: reads on the CPU with the fix done 544, of an earlier recording 1, not recorded yet 0, of this part 0; pages fixed from stamps 69463, from snapshots 1198; the shadows: 294244 pages synced from captures, 293644 made known by it, 600 left unknown because the replay or the collection wrote them since
[hos host] vk: replay reads of guest memory as submitted: 2660000 (1075971 pages from the captures)
[hos host] vk: sparse mirrors: 49, 797 MB with memory of 964 MB (166 MB spared); blocks given memory later 60, given back 1234; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 90 binds (14.2 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 7 taken from the spares, 60 made new; 18 kept as spares, 0 destroyed; 11 spares now, 296 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 814, 3411.9 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 4946 dirty pages from their submits' captures, 46413 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 866 dispatches, 37304 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6017 runs, 637.2 MB compared; 75 pages read into the shadows; 256 MB of shadows (121 MB of pages known, in 154 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 18125 in captures, 0 only in the collection, 38 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 58 live (1088 MB), 0 retired still imported (0 MB), 8 retired given back (236 MB)
[hos host] vk: replay reads of guest memory as submitted: 2680000 (1083561 pages from the captures)
[hos ModuleSystemWorker1] vk: readbacks of GPU results: 981 of 3000 waited for their write-back, 0.75 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos ModuleSystemWorker1] vk: invalidates since the last readbacks line: 3618 kept their virtual core (nothing there to wait for), 3038 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos host] vk: replay reads of guest memory as submitted: 2700000 (1091113 pages from the captures)
[hos host] vk: a frame: 60 draws kept their neighbour's descriptors (0 of them through other slots' registrations)
[hos host] vk: a frame: 0.0 draws left out (not ready), 0.00 waits for shaders or pipelines (0.0 ms); translation 3676 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 10.75, present 0.18, draw 6.54, descriptors 2.36, texture lookups 0.08, uploads 0.00, pipelines 0.19, programs 1.08, commands 2.50, dispatch 0.85, memory sync 0.30, queue submit 1.28, slot wait 0.07
[hos host] vk: a frame: mirrors 67 (797 MB): 5.41 MB synced in 1005.1 copies, 17.16 MB written back
[hos host] vk: 4080 presents, 35.0 fps; a frame: 1770 draws (0 skipped), 139 dispatches, 170 passes, 152 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.1 ms (recording slots only); 1090 pipelines (486 ms), 660 programs (11 failed), 991 textures (3554 MB)
[hos host] vk: host time in GPU resource waits: 16.17 ms a frame (fences 0.04, timelines 16.14, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 642.8 full barriers recorded, 313.6 left out (nothing recorded since the one before), 17.2 pass starts ordered in their fragment stages only (draw passes after draw passes)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 53.6; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: a frame: indirect mesh draws' arguments (RX_VK_MESH_CONV=2): 41.0 converted one by one, 0.0 from 0.0 batches at the game's barriers (0.0 converted in them; 0.0 dropped after a write, 0.0 the ring came round), 41.0 for the compute fallback (no mesh shaders in use: one by one); scans ended at a barrier 0.0, an op 0.0, the end 0.0, full 0.0; the draws batched by a scan right after an op that ended the one before, by op: none
[hos host] vk: GPU time 21.8 ms a frame (the longest submission 6.3 ms); 0.0 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.0 MB), ASTC or BC 0.0 (0.0 MB), other 0.0 (0.0 MB)
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos host] vk: the recording streams: 48 of 1 MB, the most one held 176 KB; 0 made on demand (0 MB kept now); c1 stages: at most 121 a recording; 0 descriptor pools made when a recording's own ran out (0 slots' pools made whole); a recording's pool took at most 138 sets (ub 1647, sb 917, cis 9267, si 0, utb 0, stb 0)
[hos host] vk: write-backs left out (600 submits): 836, 3504.2 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 5036 dirty pages from their submits' captures, 47495 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 882 dispatches, 38236 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6076 runs, 651.2 MB compared; 75 pages read into the shadows; 256 MB of shadows (121 MB of pages known, in 153 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 18618 in captures, 0 only in the collection, 37 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 2720000 (1098735 pages from the captures)
[hos sead::MainThread] nvn: GPU waits in 5 s: 175 sync waits (1933 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.02, writes collected 0.03
[hos host] vk: replay reads of guest memory as submitted: 2740000 (1106223 pages from the captures)
[hos Presentation Thread] vk: capture: a submit 0.10 ms (lock 0.00, scan 0.02 over 1308 ranges, 1864 MB), 217 pages written, 217 copied (0.8 MB)
[hos host] vk: a capture taken (ms): all 0.016; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.002, bytes copied 0.000; 25 kept (132 MB), 11 spare buffers, 0 texture pages copied (0 MB)
[hos host] vk: program picks: 3057022 kept (1625124 by the last picks' textures, a uniform buffer bound since), 20184 new, 1322793 with other textures (of 4400000)
[hos host] vk: sparse mirrors: 49, 797 MB with memory of 964 MB (166 MB spared); blocks given memory later 60, given back 1234; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 90 binds (14.2 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 7 taken from the spares, 60 made new; 18 kept as spares, 0 destroyed; 11 spares now, 296 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 814, 3411.9 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 4922 dirty pages from their submits' captures, 46511 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 863 dispatches, 37360 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 5971 runs, 637.5 MB compared; 75 pages read into the shadows; 256 MB of shadows (121 MB of pages known, in 154 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 18238 in captures, 0 only in the collection, 38 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 58 live (1088 MB), 0 retired still imported (0 MB), 8 retired given back (236 MB)
[hos host] vk: replay reads of guest memory as submitted: 2760000 (1113823 pages from the captures)
[hos host] vk: readback pages refreshed from guest memory: 949077 pages so far (0 held: may be written ahead in their submission)
[hos BaseProcExecutor1] vcores: 62366 takes, 7910 waits (mean 77 us, longest 5.2 ms, 0 over HOS_VCORE_LOG_MS), 3223 kicks, 3514 yields, 0 starved grants, 0 escapes in 5 s
[hos host] vk: replay reads of guest memory as submitted: 2780000 (1121423 pages from the captures)
[hos host] vk: replay reads of guest memory as submitted: 2800000 (1129210 pages from the captures)
[hos Presentation Thread] vi: 4158 frames presented (35.0 fps over the last 5.0 s, interval 1)
[hos Presentation Thread] vi: present intervals: <15.5 ms 0, 15.5-17.8 0, 17.8-22 0, 22-30 173, 30-40 2, over 40 0
[hos host] vk: write-backs left out (600 submits): 836, 3504.2 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 5035 dirty pages from their submits' captures, 47493 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 883 dispatches, 38205 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6026 runs, 649.7 MB compared; 75 pages read into the shadows; 256 MB of shadows (121 MB of pages known, in 153 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 18613 in captures, 0 only in the collection, 37 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: shadow staging (600 takes): 0 shadow copies from staging, 0 from the capture buffer; capture buffers: 600 host-cached, 0 not (write-combined: staged)
[hos host] vk: late pages found at the replay (600 takes): 0 pages in 0 takes (RX_VK_LATE_AT_REPLAY=0: only the ones the capture found)
[hos host] vk: replay reads of guest memory as submitted: 2820000 (1136855 pages from the captures)
[hos host] vk: replay reads of guest memory as submitted: 2840000 (1144485 pages from the captures)
[hos sead::MainThread] vk: capture: a submit 0.10 ms (lock 0.00, scan 0.02 over 1308 ranges, 1864 MB), 216 pages written, 216 copied (0.8 MB)
[hos host] vk: a capture taken (ms): all 0.016; GPU wait for late pages 0.000 (0 of 600 takes), pools lock 0.000, retires and stamps 0.002, bytes copied 0.000; 25 kept (138 MB), 10 spare buffers, 0 texture pages copied (0 MB)
[hos ModuleSystemWorker1] vk: readbacks of GPU results: 982 of 3000 waited for their write-back, 0.72 ms each; 0 looked at the stopped queues first, 0 waited for their submission's last part (guest memory read in place after theirs), 0 waited for their whole submission (their pages written in place, no mirror)
[hos ModuleSystemWorker1] vk: invalidates since the last readbacks line: 3834 kept their virtual core (nothing there to wait for), 3078 gave it up (counted by 256 a thread; RX_VK_INVALIDATE_QUICK=0: none kept)
[hos sead::MainThread] nvn: CommandBufferWaitEvent #70000: event 0x1006e38000 mode 0 value 1 (now 0) (lr 0x802eb444)
[hos host] vk: sparse mirrors: 49, 797 MB with memory of 964 MB (166 MB spared); blocks given memory later 60, given back 1234; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 90 binds (14.2 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 7 taken from the spares, 60 made new; 18 kept as spares, 0 destroyed; 11 spares now, 296 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 814, 3411.9 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 4903 dirty pages from their submits' captures, 46213 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 868 dispatches, 37190 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6033 runs, 640.0 MB compared; 75 pages read into the shadows; 256 MB of shadows (121 MB of pages known, in 154 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 18073 in captures, 0 only in the collection, 38 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 58 live (1088 MB), 0 retired still imported (0 MB), 8 retired given back (236 MB)
[hos host] vk: memory: working set 6174 MB (peak 6406), private 1028 MB, staging 32 MB in 2 arenas, 21 made so far (59 ms)
[hos host] vk: memory heaps: heap 0 (device): 105 blocks 5040 MB, 14174 allocations 4796 MB; used 5074 of 7109 MB; heap 1 (system): 18 blocks 600 MB, 39 allocations 542 MB; used 1764 of 15773 MB; heap 2 (device): 57 blocks 141 MB, 149 allocations 125 MB; used 264 of 256 MB;
[hos host] vk: image blocks: 1 blocks 32 MB, 6 images 31 MB; 3 free ranges, the largest 0 MB
[hos host] vk: image blocks (render targets): 18 blocks 288 MB, 95 images 152 MB; 79 free ranges, the largest 16 MB
[hos host] vk: image blocks (textures, compacted): 32 blocks 1024 MB, 882 images 961 MB; 426 free ranges, the largest 4 MB
[hos host] vk: render target images: 234 reused from the pool, 245 made new, 147 destroyed from it, 11 reused by size (on a pooled one's memory, RX_VK_TARGET_REUSE_MEMORY=1), 2 missed after their image was taken by size; it holds 0 (0 MB)
[hos host] vk: render targets not reused by size: 180 with the pool empty (0 of them with targets still in the dead queue), 6 with none in it a present yet, 0 with none of a kind they can use, 52 with all smaller (the largest 48% short on average), 3 with all bigger (over 1.5 times), 4 with some of each, 0 over a block, 0 refused by the driver, 0 not bound; the pool let go of 0 over its bound and 147 idle
[hos host] vk: VRAM: 5338 MB used of 7365 MB allowed (259 MB of it free in the blocks); textures 3554 MB (budget 4629 MB), 992 textures
[hos host] vk: budget: VRAM 5181 of 7365 MB (5338 MB by the driver, 157 MB unaccounted): textures 3554, mirrors 797, blocks free 259, shadows 256, other buffers 201, other images 79, staging 32; RAM 1657 MB (the process's private memory 1028 MB): imports 1088, capture spares 400, captures 128, capture table 27, other buffers 14, others 0
[hos host] vk: textures: decoded ASTC 1125 MB, rendered 656 MB (78 drawn in one frame, kept: 41805 KB), other 245 MB; used in the last 8 presents 1964 MB
[hos host] vk: decoded ASTC used again: 0 textures (0 MB) copied from ones the game finalized, not decoded; 0 of those retired, 0 listed; 503 MB of blocks hashed in 46 ms
[hos host] vk:   288 MB: 0x1147c16000 12288x12288x1 x1 levels 1 NVN 0x32 rendered, last used 1 presents ago
[hos host] vk:   214 MB: 0x10a22aa600 1365x1365x1 x90 levels 11 NVN 0x8e ASTC, last used 1 presents ago
[hos host] vk:   214 MB: 0x10a5b22c00 1365x1365x1 x90 levels 11 NVN 0x80 ASTC, last used 1 presents ago
[hos host] vk:   64 MB: 0x113755c000 4096x4096x1 x1 levels 1 NVN 0x34 rendered, last used 1 presents ago
[hos host] vk:   64 MB: 0x113cba9000 512x512x64 x1 levels 1 NVN 0x3d rendered, last used 1 presents ago
[hos host] vk:   47 MB: 0x10c2c99000 3072x3072x1 x1 levels 12 NVN 0x7b ASTC, last used 3 presents ago
[hos host] vk:   47 MB: 0x10b4239e00 3072x3072x1 x1 levels 12 NVN 0x7b ASTC, last used 1 presents ago
[hos host] vk:   47 MB: 0x10b4ad6c00 3072x3072x1 x1 levels 12 NVN 0x7b ASTC, last used 1 presents ago
[hos host] vk: finalized textures the GPU drew once, kept for being made again: 30 (9229 KB): 8x8x1 NVN 0x3f: 1 (2 KB); 62x41x1 NVN 0x25: 1 (9 KB); 74x32x1 NVN 0x25: 1 (9 KB); 90x34x1 NVN 0x25: 1 (11 KB); 126x105x1 NVN 0x1: 1 (12 KB); 138x96x1 NVN 0x1: 1 (12 KB); 154x98x1 NVN 0x1: 1 (14 KB); 198x40x1 NVN 0x25: 2 (61 KB); 242x40x1 NVN 0x25: 3 (113 KB); 260x260x1 NVN 0xd: 1 (175 KB); 262x104x1 NVN 0x1: 3 (79 KB); 306x104x1 NVN 0x1: 3 (93 KB); 592x49x1 NVN 0x25: 3 (339 KB); 656x113x1 NVN 0x1: 3 (217 KB); 1466x282x1 NVN 0x25: 5 (8074 KB);
[hos host] vk: a submit's memory work (ms): maps 0.00, capture taken 0.02, writes collected 0.03
[hos Presentation Thread] vk: presents after the CPU's signal (600): 600 waited for it, 11.29 ms each, 0 let go after RX_VK_FRAME_GATE_MS
[hos host] vk: status: 600 frames, the slowest 84.7 ms; VRAM 5.2 of 7.2 GB
[hos host] vk: program census (play): created 0 (0 with stages; 0 with no present for 100 ms before, 0 in a frame of 100 ms or more; 0.0 ms of the game's threads), freed 0 (0 never used); first uses 0: program ready 0, loaded 0, translated 0, failed 0; pipeline ready 0, from a cache 0, compiled 0, made 0, left out 0, none 0; 0 with no create call seen, 0 whose stages are not the create call's; after the create call: under 16.7 ms 0, under 100 ms 0, under 1 s 0, longer 0; needing work 0 (0 of them 100 ms or more after the create call, 0 created in a load or a long frame); in play: 0 program records loaded (0.0 ms), 813 new graphics pipelines, pipelines from the driver's cache 993, from the start's cache 0, compiled 0; 2882 draws and dispatches left out, 562 waits (2816.0 ms)
[hos host] vk: program loads at create (play): 0 queued, 0 loaded (median 0.00 ms, 99th percentile 0.00 ms, worst 0.00 ms, 0 ms and 0.0 MB read in all); 0 programs (0 of them compute variants without their code, 0 graphics programs no note names), 0 graphics pipelines (0 not in a cache), 0 compute pipelines (0 not in a cache); a first use ran 0 itself and waited for 0 (0.0 ms); 0 pipelines made at create taken by draws; 7015 create calls with nothing kept for their stages, 0 before the index; units 7015 (0 ready, 0 loading), 12472 programs alive; 0 noted graphics pipeline probes deferred to first use
[hos host] vk: program loads at create (play): 0 shader modules let go once their programs' noted pipelines were built, 0 programs' modules made again for a new pipeline state (RX_VK_DROP_LOADED_MODULES=0 keeps them)
[hos host] vk: module identifiers (play): 0 known, 0 learned in this run; 0 compute pipelines made by them at a create call, 0 asked that way were in no cache (their code read, as before)
[hos host] vk: a frame: 60 draws kept their neighbour's descriptors (0 of them through other slots' registrations)
[hos host] vk: a frame: 0.0 draws left out (not ready), 0.00 waits for shaders or pipelines (0.0 ms); translation 3676 ms in all, 0 jobs queued
[hos host] vk: a frame's replay time (ms): submit 10.69, present 0.18, draw 6.52, descriptors 2.37, texture lookups 0.08, uploads 0.00, pipelines 0.19, programs 1.07, commands 2.50, dispatch 0.86, memory sync 0.29, queue submit 1.26, slot wait 0.07
[hos host] vk: a frame: mirrors 67 (797 MB): 5.41 MB synced in 1003.8 copies, 17.20 MB written back
[hos host] vk: 4200 presents, 35.0 fps; a frame: 1770 draws (0 skipped), 139 dispatches, 170 passes, 152 barriers, 6 submits, 0.0 uploads (0.0 MB), 80 ub updates (56 overlays), GPU-bound wait 0.1 ms (recording slots only); 1090 pipelines (486 ms), 660 programs (11 failed), 992 textures (3554 MB)
[hos host] vk: host time in GPU resource waits: 16.15 ms a frame (fences 0.04, timelines 16.11, idle 0.00, queries 0.00; sum across threads, not frame time; excludes pacing and driver submit/present blocking)
[hos host] vk: a frame: 643.0 full barriers recorded, 313.6 left out (nothing recorded since the one before), 17.2 pass starts ordered in their fragment stages only (draw passes after draw passes)
[hos host] vk: a frame: the game's barriers by their NVN flags: 0x1: 2.0; 0x8: 3.0; 0x9: 1.0; 0x11: 53.6; 0x12: 68.5; 0x13: 2.0; 0x21: 4.0; 0x28: 14.0; 0x92: 3.0; 0x97: 1.0;
[hos host] vk: a frame: indirect mesh draws' arguments (RX_VK_MESH_CONV=2): 41.0 converted one by one, 0.0 from 0.0 batches at the game's barriers (0.0 converted in them; 0.0 dropped after a write, 0.0 the ring came round), 41.0 for the compute fallback (no mesh shaders in use: one by one); scans ended at a barrier 0.0, an op 0.0, the end 0.0, full 0.0; the draws batched by a scan right after an op that ended the one before, by op: none
[hos host] vk: GPU time 21.8 ms a frame (the longest submission 6.3 ms); 0.0 textures made a frame: colour targets 0.0 (0.0 MB), depth 0.0 (0.0 MB), ASTC or BC 0.0 (0.0 MB), other 0.0 (0.0 MB)
[hos host] vk: a frame: 0.00 MB unswizzled on the CPU in 0.00 ms
[hos host] vk: the render thread (600 presents, 28.93 ms a frame): its jobs 11.83 ms a frame (submits 1.07, presents 0.18, others 10.58), idle 17.10; the game waited for room at 0 presents, 0.00 ms each (0.00 a frame; 1.00 jobs queued at a present); 0 catch-ups (a finish, a texture read back, a write-back not recorded yet: 0), 0.00 ms each; 0 presents held for the CPU's signal, 0.00 ms each (0 let go by RX_VK_HOLD_MS); 600 captures at a CPU signal; the submits' waits read in 0.036 ms a frame; readbacks: 3000 next to a texture's storage (no catch-up), 0 in one, 0 read exactly (after the work handed over: among the catch-ups); a present made 10.62 ms after its hand-over (58.32 at most)
[hos host] vk: frame overlap (600 presents): the GPU between recordings, ms a frame, idle + in the next one's uploads: up to a frame's first recording with a draw 0.10 + 0.90 (the window's copy is in the idle), after it 1.09 + 5.00; after a frame's last recording that reads or writes the game's memory come 1.00 recordings (0.22 ms of GPU time), after its last that reads it 1.00 (0.22 ms)
[hos host] vk: frame ahead (600 presents): 8.96 parts a frame recorded in the signal's job, 66.1% of them took nothing that came with the signal; of the others, 3.04 a frame took pages of the signal's capture into a mirror (133.0 pages a frame) and 0.00 pages read as guest memory had them (0.0 pages a frame); a frame's first part took nothing in 0 of 600 frames, its first three in 0
[hos host] vk: frame ahead, what came with the signal: 600 signals with a capture, each of 165.0 pages in 10.0 pools; between a frame's last submit and that signal 0.00 texture and sampler registrations, 0.00 releases and 0.00 other CPU signals a signal (0 of them with any)
[hos host] vk: frame overlap, by place: idle + uploads before a frame's 2nd to 12th recording with a draw, then all later ones: 0.84+2.35, 0.03+0.40, 0.02+1.73, 0.02+0.09, 0.13+0.09, 0.01+0.08, 0.01+0.11, 0.02+0.08, 0.01+0.08, 0.00+0.00, 0.00+0.00, 0.00+0.00
[hos ModuleSystemWorker4] nvn: CommandBufferSignalEvent #50000: event 0x100d6c8010 mode 0 value 50002 (lr 0x80660bcc)
[hos host] vk: replay reads of guest memory as submitted: 2860000 (1152573 pages from the captures)
[hos host] vk: render targets drawn over from another start, let go: 318 so far (474 MB), 307 of them (457 MB) between the passes
[hos ModuleSystemWorker4] tex: the game's texture heap: 89 compaction rounds asked for in 3600 frames without one running
[hos host] vk: program picks: 3197174 kept (1699684 by the last picks' textures, a uniform buffer bound since), 20293 new, 1382532 with other textures (of 4600000)
[hos host] vk: replay reads of guest memory as submitted: 2880000 (1160630 pages from the captures)
[hos host] vk: a submission's parts ended: 15100 where the game had ended its render pass, 4900 at 512 draws
[hos host] vk: write-backs left out (600 submits): 836, 3504.2 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 5024 dirty pages from their submits' captures, 48019 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 886 dispatches, 38532 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6058 runs, 651.3 MB compared; 76 pages read into the shadows; 256 MB of shadows (121 MB of pages known, in 154 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 18698 in captures, 0 only in the collection, 38 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: replay reads of guest memory as submitted: 2900000 (1170598 pages from the captures)
[hos host] vk: replay reads of guest memory as submitted: 2920000 (1178712 pages from the captures)
[hos host] vk: slow submit: 36.1 ms: dispatch 28.8
[hos host] vk: late pages' fix: stamps for 0x1051b90000 (pool 0x1051800000; 448 blocks, 28672 KB)
[hos host] vk: late pages' fix: stamps for 0x1137390000 (pool 0x1137000000; 512 blocks, 32768 KB)
[hos host] vk: late pages' fix: stamps for 0x1137790000 (pool 0x1137000000; 576 blocks, 36864 KB)
[hos sead::MainThread] nvn: GPU waits in 5 s: 171 sync waits (1979 ms), 0 polls timed out (0 ms), 0 queue finishes (0 ms)
[hos host] vk: the recording streams: 48 of 1 MB, the most one held 176 KB; 0 made on demand (0 MB kept now); c1 stages: at most 121 a recording; 0 descriptor pools made when a recording's own ran out (0 slots' pools made whole); a recording's pool took at most 138 sets (ub 1647, sb 917, cis 9267, si 0, utb 0, stb 0)
[hos host] vk: sparse mirrors: 49, 797 MB with memory of 964 MB (166 MB spared); blocks given memory later 60, given back 1234; bindings served from guest memory meanwhile 0 (0 reaching blocks with memory too), from the mirror past every game pool 0; 90 binds (14.2 ms); 0 made whole for want of memory
[hos host] vk: sparse mirrors' buffers: 7 taken from the spares, 60 made new; 18 kept as spares, 0 destroyed; 11 spares now, 296 MB of address space (RX_VK_SPARSE_SPARE_MB=4096)
[hos host] vk: write-backs left out (600 submits): 836, 3504.2 MB; 23 bindings followed, 34 ranges the CPU read, 0 resets; CPU reads or writes of one left out: 0 so far
[hos host] vk: mirror syncs (600 submits): 4914 dirty pages from their submits' captures, 47658 from earlier ones', 0 from guest memory as it was at the replay; 25 captures kept; 0 pages the GPU wrote copied over (0 live); 0 kept for the GPU (RX_VK_PROTECT_BARE)
[hos host] vk: mirror syncs' pass (600 submits): 881 dispatches, 38342 chunks; 0 syncs as copies (no table room)
[hos host] vk: write-back diff (600 submits): 6067 runs, 651.5 MB compared; 199 pages read into the shadows; 256 MB of shadows (121 MB of pages known, in 154 MB of 1 MB parts); pages the shadows knew that the CPU wrote: 18533 in captures, 0 only in the collection, 39 only by the replay itself
[hos host] vk:   read live, by pool:
[hos host] vk: imports: 58 live (1088 MB), 0 retired still imported (0 MB), 8 retired given back (236 MB)
[hos sead::MainThread] nvn: 450000 recordings made, 48971 live copies (16.3 MB), 370 texture copies; 7 command buffers' segments (4.4 MB)
[hos host] vk: replay reads of guest memory as submitted: 2940000 (1186447 pages from the captures)
[hos Presentation Thread] vk: capture: a submit 0.10 ms (lock 0.00, scan 0.02 over 1308 ranges, 1864 MB), 215 pages written, 215 copied (0.8 MB)
[hos host] vk: a capture taken (ms): all 0.045; GPU wait for late pages 0.029 (1 of 600 takes), pools lock 0.000, retires and stamps 0.002, bytes copied 0.000; 24 kept (130 MB), 12 spare buffers, 0 texture pages copied (0 MB)
[hos host] control: quit asked by a client
[hos host] vk: program census (close): created 0 (0 with stages; 0 with no present for 100 ms before, 0 in a frame of 100 ms or more; 0.0 ms of the game's threads), freed 0 (0 never used); first uses 0: program ready 0, loaded 0, translated 0, failed 0; pipeline ready 0, from a cache 0, compiled 0, made 0, left out 0, none 0; 0 with no create call seen, 0 whose stages are not the create call's; after the create call: under 16.7 ms 0, under 100 ms 0, under 1 s 0, longer 0; needing work 0 (0 of them 100 ms or more after the create call, 0 created in a load or a long frame); in play: 0 program records loaded (0.0 ms), 813 new graphics pipelines, pipelines from the driver's cache 993, from the start's cache 0, compiled 0; 2882 draws and dispatches left out, 562 waits (2816.0 ms)
[hos host] vk: program loads at create (close): 0 queued, 0 loaded (median 0.00 ms, 99th percentile 0.00 ms, worst 0.00 ms, 0 ms and 0.0 MB read in all); 0 programs (0 of them compute variants without their code, 0 graphics programs no note names), 0 graphics pipelines (0 not in a cache), 0 compute pipelines (0 not in a cache); a first use ran 0 itself and waited for 0 (0.0 ms); 0 pipelines made at create taken by draws; 7015 create calls with nothing kept for their stages, 0 before the index; units 7015 (0 ready, 0 loading), 12472 programs alive; 0 noted graphics pipeline probes deferred to first use
[hos host] vk: program loads at create (close): 0 shader modules let go once their programs' noted pipelines were built, 0 programs' modules made again for a new pipeline state (RX_VK_DROP_LOADED_MODULES=0 keeps them)
[hos host] vk: module identifiers (close): 0 known, 0 learned in this run; 0 compute pipelines made by them at a create call, 0 asked that way were in no cache (their code read, as before)
[hos host] vk: indexed descriptor arrays (close): 42 in 42 of 1304 program stages, 0 of them near a limit
[hos host] vk: indexed descriptor arrays (close), fragment stages: 42 texture (17-64: 42); the longest 64
[hos host] vk: page table census (close): no pool outside its windows, 64-72 GB and 508-512 GB (4524 pools in the table now)
[hos host] control: exiting
[hos host] stop: close work done (a file the game was writing waited for, at most 3 s)